In 2009 I posted an admittedly long-shot theory about suicidality in rodents and maladaptive (violent) behavior in humans, namely serial murder. Along comes a very interesting piece in Archives of General Psychiatry by Pedersen et al, Toxoplasma gondii Infection and Self-directed Violence in Mothers.
Toxoplasma is that bug that infects rodents and somehow makes them approach cats. The parasite then goes through the next stage of its life cycle in the cat's gut, and is defecated out, at which point it infects another rodent. Another long-shot but interesting possibility is that the "cat lady" phenomenon is actually a toxoplasmosis infection: if you're an animal so big that the cats can't eat you, you just know you want to be around cats, so you get more and more. We do know already there are definite effects in humans, although so far they could just be explained as boring motor retardation effects: in a U.S. military study, people who totaled their vehicles in accidents were significantly more likely to have Toxo antibodies.
I think we're eventually going to learn that Toxoplasma is able to manipulate the behavior of humans in very specific uncomfortable ways vis a vis rabies, which is itself an amazing pathogen. Humans and tox have a long history together. In fact, the Rh factor on red blood cells is a subunit of an ammonia channel that's implicated in resistance to tox, and it's not surprising that the only part of the world with appreciable Rh negativity is Europe (where until recent times there wasn't much cat feces around). The closer you get to Basque country, the higher the Rh negativity; interesting since the Basques have been in (barely) post-glacial Europe longer than any other population. For a system that could result in immunized Rh- mothers that after their first child can produce very sick anemic kids, it has to be doing something important.